Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Deep-Learning Channel Estimation for IRS-Assisted Integrated Sensing and Communication System

Yu Liu, Ibrahim Al-Nahhal, Octavia A. Dobre, Fanggang Wang

TL;DR

This work tackles channel estimation in IRS-assisted ISAC systems, where a passive IRS and SAC interference complicate estimation. It introduces a three-stage, CNN-based framework that decouples direct SAC channels $\mathbf b$ and $\mathbf f$ from reflected channels $\mathbf G_{\mathrm u}$ and $\mathbf G_{\mathrm t}$, using stage-specific pilot designs and two CNN architectures (DE-CNN and RE-CNN). The approach shows substantial NMSE improvements over LS baselines across SNRs and system dimensions, with up to 12.5 dB SNR gains in key channels and good generalization, while incurring acceptable computational complexity. These results demonstrate practical viability for accurate channel estimation in IRS-enhanced ISAC networks, enabling improved sensing and communication performance through learned, stage-wise channel mappings.

Abstract

Integrated sensing and communication (ISAC), and intelligent reflecting surface (IRS) are envisioned as revolutionary technologies to enhance spectral and energy efficiencies for next wireless system generations. For the first time, this paper focuses on the channel estimation problem in an IRS-assisted ISAC system. This problem is challenging due to the lack of signal processing capacity in passive IRS, as well as the presence of mutual interference between sensing and communication (SAC) signals in ISAC systems. A three-stage approach is proposed to decouple the estimation problem into sub-ones, including the estimation of the direct SAC channels in the first stage, reflected communication channel in the second stage, and reflected sensing channel in the third stage. The proposed three-stage approach is based on a deep-learning framework, which involves two different convolutional neural network (CNN) architectures to estimate the channels at the full-duplex ISAC base station. Furthermore, two types of input-output pairs to train the CNNs are carefully designed, which affect the estimation performance under various signal-to-noise ratio conditions and system parameters. Simulation results validate the superiority of the proposed estimation approach compared to the least-squares baseline scheme, and its computational complexity is also analyzed.

Deep-Learning Channel Estimation for IRS-Assisted Integrated Sensing and Communication System

TL;DR

This work tackles channel estimation in IRS-assisted ISAC systems, where a passive IRS and SAC interference complicate estimation. It introduces a three-stage, CNN-based framework that decouples direct SAC channels and from reflected channels and , using stage-specific pilot designs and two CNN architectures (DE-CNN and RE-CNN). The approach shows substantial NMSE improvements over LS baselines across SNRs and system dimensions, with up to 12.5 dB SNR gains in key channels and good generalization, while incurring acceptable computational complexity. These results demonstrate practical viability for accurate channel estimation in IRS-enhanced ISAC networks, enabling improved sensing and communication performance through learned, stage-wise channel mappings.

Abstract

Integrated sensing and communication (ISAC), and intelligent reflecting surface (IRS) are envisioned as revolutionary technologies to enhance spectral and energy efficiencies for next wireless system generations. For the first time, this paper focuses on the channel estimation problem in an IRS-assisted ISAC system. This problem is challenging due to the lack of signal processing capacity in passive IRS, as well as the presence of mutual interference between sensing and communication (SAC) signals in ISAC systems. A three-stage approach is proposed to decouple the estimation problem into sub-ones, including the estimation of the direct SAC channels in the first stage, reflected communication channel in the second stage, and reflected sensing channel in the third stage. The proposed three-stage approach is based on a deep-learning framework, which involves two different convolutional neural network (CNN) architectures to estimate the channels at the full-duplex ISAC base station. Furthermore, two types of input-output pairs to train the CNNs are carefully designed, which affect the estimation performance under various signal-to-noise ratio conditions and system parameters. Simulation results validate the superiority of the proposed estimation approach compared to the least-squares baseline scheme, and its computational complexity is also analyzed.
Paper Structure (24 sections, 42 equations, 9 figures, 1 table, 1 algorithm)

This paper contains 24 sections, 42 equations, 9 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. System Model
  3. Proposed Three-Stage Estimation Approach
  4. Design of Three-Stage Approach
  5. First Stage
  6. Second Stage
  7. Third Stage
  8. Input-Output Pairs Design
  9. First Stage
  10. Second Stage
  11. Third Stage
  12. Proposed CNN-Based DL Framework
  13. Training Dataset Generation
  14. Offline Training
  15. Online Estimation
  16. ...and 9 more sections

Figures (9)

  • Figure 1: IRS-assisted ISAC MISO system model.
  • Figure 2: Pilot transmission protocol.
  • Figure 3: The proposed CNN-based DL framework: (a) Offline training, (b) Online estimation.
  • Figure 4: Simulation setup.
  • Figure 5: MSE of CNNs training and testing versus epoch for $M=4$, $L=15$, and $\text{SNR}=15\rm\,dB$: (a) DE-CNN for ${\mathrm S}_1$, (b) RE-CNN for ${\mathrm S}_2$, (c) RE-CNN for ${\mathrm S}_3$.
  • ...and 4 more figures